The objective of this application is to investigate how the brain ensures that rapid eye movements, called saccades, land accurately on objects of interest. Because saccades remain accurate until we are 70 years old, the brain must be capable of adjusting the efficacy of its saccade generating circuits to compensate for the general weakening of muscle and deterioration of neurons with aging. Our hypothesis is that the cerebellum, in particular its caudal fastigial nucleus (CFN), plays an important role in such continual motor adaptation or learning. We suggest that the saccadic command to the brain-stem burst generator for saccades originates in the superior colliculus (SC) and that the CFN refines this signal to produce an accurate saccade. We will test this suggestion by two sets of experiments. To explore the relative contributions of the SC and the CFN in the production of accurate saccades, we will: (1) determine the effect of inactivating the CFN on saccades generated by SC stimulation; (2) determine the effect of inactivating the nucleus reticularis tegmenti pontis (NRTP), the major input to the CFN, on saccade metrics; and (3) determine whether the CFN also participates in rapid gaze shifts, a combination of eye and head movements with the head free, by both recording neuronal activity from and by inactivating the CFN. To explore the roles of the SC and CFN in maintaining saccade accuracy, we will cause a trained monkey to undergo a behavioral change in its saccadic accuracy by surreptitiously stepping a target forward or backward as he makes a saccade toward it. After repeated trials in this situation, the animal's saccades gradually become dysmetric, either overshooting or undershooting the target, respectively. As the animal undergoes this adaptation, which takes ~1/2 hour, we will: (4) determine whether movement fields in the SC are altered; (5) determine the effect of inactivation of NRTP; and (6) determine the effect on the discharge of Purkinje cells, cerebellar interneurons that powerfully inhibit CFN neurons. Because monkey saccades are remarkably similar to ours, accomplishing these scientific aims allows us to learn more about the human condition. For example, the similarity of the dysmetria produced in Wallenberg's disease with that produced by CFN inactivation has enabled clinicians to implicate the cerebellum in that disease. Also, revealing how the brain learns to make adjustments in saccade accuracy may help us understand how the nervous system compensates for the changes that accompany growth, aging and injury for other movements.